\begin{table}
\centering
\def\sym#1{\ifmmode^{#1}\else\(^{#1}\)\fi}
\def\onepc{$^{\ast\ast}$} \def\fivepc{$^{\ast}$}
\def\zeropc{$^{\ast\ast\ast}$}
\footnotesize
\caption{Differences in Coefficient Size (Seemingly Unrelated Models)  \label{tab:sureg}}
\begin{tabular}{l*{6}{l}}
\hline
\hline
  &QVSR  &Likert+  &QVSR  &QVSR &Likert+  &QVSR  \\
  &vs. & vs. & vs. & vs. & vs. & vs.  \\
  & Likert& Likert&LIkert+&Likert& Likert& LIkert+\\
 \hline 
Gun         &     0.59***&     0.20*&     0.38***&&&\\
&     0.14&     0.10&     0.14&&&\\
Immigration &     0.37***&     0.19*&     0.17&&&\\
 &     0.13&     0.10&     0.14&&&\\
Minimum Wage&&&&     0.04&     0.08&     0.05\\
&&&&     0.24&     0.19&     0.28\\
Abortion    &&&&     0.68***&     0.38**&     0.36*\\
&&&&     0.21&     0.17&     0.22\\
DG Punish (1)&&&&     0.46&     0.39&    -0.06\\
&&&&     0.34&     0.33&     0.35\\
DG Punish (2)&&&&     0.98***&     0.37&     0.49\\
&&&&     0.37&     0.33&     0.37\\
 \hline 
N           &    2,409&    2,512&    2,417&      976&    1,017&      995\\
F test      &    28.14&     7.87&     8.24&     8.49&     3.18&     1.21\\
Prob > F    &    0.000&    0.005&    0.004&    0.004&    0.075&    0.271\\
\hline
\hline
\end{tabular}
\begin{tablenotes}[online]
\begin{scriptsize}
\item  \fivepc $p$\textless.05, \onepc $p$\textless.01 \zeropc $p$\textless.001. We replicate Figure 3 analysis using seemingly unrelated models. This table reports the interaction between the preference variable and a dummy variable identifying the surve
> y methods used. For example, for the gun donation outcome, the difference between the coefficient for Likert and that for QVSR is equal to 0.59. Bottom row: F-test for the null-hypothesis that the sum of the coefficients is equal to 0.
\end{scriptsize}
\end{tablenotes}
\end{table}
